cifar10_training_singleGPU.ipynb

{
 "metadata": {
  "language": "lua",
  "name": "",
  "signature": "sha256:6fa560d0241200804bd783d928ea0ddf53afb0c0e530e90394c1926cd21003a0"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "require \"cudnn\"\n",
      "require \"cunn\"\n",
      "cutorch.setDevice(2)\n",
      "Plot = require \"itorch.Plot\"\n",
      "utils = require \"cifar10_utils\"\n",
      "data = require(\"cifar10_load\")\n",
      "trainset = data.trainset:normalize()\n",
      "testset = data.testset:normalize()\n",
      "classes = data.classes\n",
      "collectgarbage()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "model_name = \"all-cnn-c-normalized-multyGPU\"\n",
      "load_model_name = model_name\n",
      "--load_model_name = \"vgg_10cw-5mp-3ls-N-cu\""
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "load_trained = true\n",
      "if load_trained == false then\n",
      "    utils.net = require(\"models/\" .. load_model_name)\n",
      "else\n",
      "    utils.net = torch.load(\"trained_models/\"..load_model_name)\n",
      "end\n",
      "\n",
      "utils.criterion = nn.ClassNLLCriterion():cuda()\n",
      "utils.net:cuda()\n",
      "\n",
      "utils.parameters, utils.gradParameters = utils.net:getParameters()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "--[[\n",
      "local n = torch.load(\"trained_models/fconv\")\n",
      "local lst1 = {}\n",
      "local size = 2\n",
      "for i = 1, size do\n",
      "    if n:get(i).weight then\n",
      "      lst1[#lst1 +1] = i \n",
      "    end\n",
      "end\n",
      "local lst2 = {}\n",
      "for i = 1, size do\n",
      "    if utils.net:get(i).weight then\n",
      "      lst2[#lst2 +1] = i \n",
      "    end\n",
      "end\n",
      "\n",
      "for i = 1, math.min(#lst1, #lst2) do\n",
      "   utils.net:get(lst2[i]).weight:copy(n:get(lst1[i]).weight)\n",
      "end\n",
      "    \n",
      "collectgarbage() "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "ename": "[string \"--[[...\"]:22: unfinished long comment near '<eof>'",
       "evalue": "",
       "output_type": "pyerr",
       "traceback": [
        "[string \"--[[...\"]:22: unfinished long comment near '<eof>'"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "utils.augment = utils.no_augment"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "log = {}\n",
      "--local train_loss, train_hist, train_accu = utils:evaluate(trainset)\n",
      "--local test_loss, test_hist, test_accu = utils:evaluate(testset)\n",
      "--log.train_losses = {train_loss}\n",
      "--log.train_accu = {train_accu}\n",
      "--log.test_losses = {test_loss}\n",
      "--log.test_accu = {test_accu}\n",
      "log.train_losses = {2.3}\n",
      "log.train_accu = {0.9}\n",
      "log.test_losses = {2.3}\n",
      "log.test_accu = {0.9}\n",
      "log.epoch = {0}"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "utils.opt.optimState.batch_size = 64\n",
      "utils.opt.optimState.learningRate = 1e-4\n",
      "utils.opt.optimState.weightDecay = 5e-3\n",
      "utils.opt.optimState.momentum = 0.9"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "local timer = torch.Timer()\n",
      "n_epoch = 2\n",
      "print_epoch = 1\n",
      "utils.err.any = false\n",
      "\n",
      "local loss_plot = Plot()\n",
      "loss_plot:title(\" 0 - 0% - 0 minutes\")\n",
      "loss_plot:line(log.epoch, log.train_losses, \"red\")\n",
      "loss_plot:line(log.epoch, log.train_accu, \"purple\")\n",
      "loss_plot:line(log.epoch, log.test_losses, \"blue\")\n",
      "loss_plot:line(log.epoch, log.test_accu, \"green\"):draw()\n",
      "\n",
      "for i = 1, n_epoch do\n",
      "    utils:train_epoch()    \n",
      "    if utils.err.any == true then\n",
      "        print(\"stoping because there was an error\")\n",
      "        break\n",
      "    end\n",
      "    if i % print_epoch == 0 then\n",
      "      local train_loss, train_hist, train_accu = utils:evaluate(trainset)\n",
      "      local test_loss, test_hist, test_accu = utils:evaluate(testset)\n",
      "      table.insert(log.train_losses, train_loss)\n",
      "      table.insert(log.test_losses, test_loss)\n",
      "      table.insert(log.train_accu, train_accu)\n",
      "      table.insert(log.test_accu, test_accu)\n",
      "      table.insert(log.epoch, log.epoch[#log.epoch] + print_epoch)\n",
      "      print(\"------------------------------------------------\")\n",
      "      print(\"epoch \" .. log.epoch[#log.epoch] .. \":\")\n",
      "      print(string.format(\" train loss = %.4f  -  train accuracy = %.2f%%\", train_loss, 100 - 100*train_accu))\n",
      "      print(string.format(\" test  loss = %.4f  -  test  accuracy = %.2f%%\", test_loss, 100 - 100*test_accu))\n",
      "      loss_plot:title(string.format(\" %i - %.0f%% - %.0f minutes\", i, i/n_epoch*100, timer:time().real/60))\n",
      "      loss_plot:line(log.epoch, log.train_losses, \"red\")\n",
      "      loss_plot:line(log.epoch, log.train_accu, \"purple\")\n",
      "      loss_plot:line(log.epoch, log.test_losses, \"blue\")\n",
      "      loss_plot:line(log.epoch, log.test_accu, \"green\"):redraw()\n",
      "      utils:ping()\n",
      "    end\n",
      "end\n",
      "collectgarbage()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<script type=\"text/javascript\">\n",
        "$(function() {\n",
        "    if (typeof (window._bokeh_onload_callbacks) === \"undefined\"){\n",
        "  window._bokeh_onload_callbacks = [];\n",
        "    }\n",
        "    function load_lib(url, callback){\n",
        "  window._bokeh_onload_callbacks.push(callback);\n",
        "  if (window._bokeh_is_loading){\n",
        "      console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", new Date());\n",
        "      return null;\n",
        "  }\n",
        "  console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", new Date());\n",
        "  window._bokeh_is_loading = true;\n",
        "  var s = document.createElement('script');\n",
        "  s.src = url;\n",
        "  s.async = true;\n",
        "  s.onreadystatechange = s.onload = function(){\n",
        "      Bokeh.embed.inject_css(\"http://cdn.pydata.org/bokeh-0.7.0.min.css\");\n",
        "      window._bokeh_onload_callbacks.forEach(function(callback){callback()});\n",
        "  };\n",
        "  s.onerror = function(){\n",
        "      console.warn(\"failed to load library \" + url);\n",
        "  };\n",
        "  document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
        "    }\n",
        "\n",
        "    bokehjs_url = \"http://cdn.pydata.org/bokeh-0.7.0.min.js\"\n",
        "\n",
        "    var elt = document.getElementById(\"3de3aa81-2bc2-4514-c55b-552a597be5ee\");\n",
        "    if(elt==null) {\n",
        "  console.log(\"Bokeh: ERROR: autoload.js configured with elementid '3de3aa81-2bc2-4514-c55b-552a597be5ee'\"\n",
        "        + \"but no matching script tag was found. \")\n",
        "  return false;\n",
        "    }\n",
        "\n",
        "    if(typeof(Bokeh) !== \"undefined\") {\n",
        "  console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n",
        "  var modelid = \"fb8d1727-b786-45ce-c531-9c54b5609220\";\n",
        "  var modeltype = \"Plot\";\n",
        "  var all_models = [{\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[2.3],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\",\"doc\":null,\"tags\":[]}},{\"id\":\"294bc1d6-012f-4c1b-ccc7-8ff65da8cfb9\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"red\"},\"line_color\":{\"value\":\"red\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"294bc1d6-012f-4c1b-ccc7-8ff65da8cfb9\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"8065d0d8-e099-4afe-ca1c-cd2fa32b7fc2\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"red\"},\"line_color\":{\"value\":\"red\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"8065d0d8-e099-4afe-ca1c-cd2fa32b7fc2\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"b8bff12f-b404-43d6-c3ca-4404aedee39e\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"8065d0d8-e099-4afe-ca1c-cd2fa32b7fc2\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\"},\"glyph\":{\"type\":\"Line\",\"id\":\"294bc1d6-012f-4c1b-ccc7-8ff65da8cfb9\"},\"selection_glyph\":null,\"id\":\"b8bff12f-b404-43d6-c3ca-4404aedee39e\",\"tags\":[]}},{\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[0.9],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\",\"doc\":null,\"tags\":[]}},{\"id\":\"50ab886e-2ffc-4562-c531-022fe52b2fa8\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"purple\"},\"line_color\":{\"value\":\"purple\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"50ab886e-2ffc-4562-c531-022fe52b2fa8\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"22748664-18c5-491a-c983-aa6519a2cf3d\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"purple\"},\"line_color\":{\"value\":\"purple\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"22748664-18c5-491a-c983-aa6519a2cf3d\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"dfc5d210-21c2-4941-c9e8-e315d3e43f11\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"22748664-18c5-491a-c983-aa6519a2cf3d\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\"},\"glyph\":{\"type\":\"Line\",\"id\":\"50ab886e-2ffc-4562-c531-022fe52b2fa8\"},\"selection_glyph\":null,\"id\":\"dfc5d210-21c2-4941-c9e8-e315d3e43f11\",\"tags\":[]}},{\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[2.3],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\",\"doc\":null,\"tags\":[]}},{\"id\":\"5009ef23-82c9-4a61-ca83-74682f17fda2\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"blue\"},\"line_color\":{\"value\":\"blue\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"5009ef23-82c9-4a61-ca83-74682f17fda2\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"cb31dc76-94d6-465b-c6f7-be2f9280a00e\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"blue\"},\"line_color\":{\"value\":\"blue\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"cb31dc76-94d6-465b-c6f7-be2f9280a00e\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"2ce6d563-645b-437e-c3e5-955c0880629e\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"cb31dc76-94d6-465b-c6f7-be2f9280a00e\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\"},\"glyph\":{\"type\":\"Line\",\"id\":\"5009ef23-82c9-4a61-ca83-74682f17fda2\"},\"selection_glyph\":null,\"id\":\"2ce6d563-645b-437e-c3e5-955c0880629e\",\"tags\":[]}},{\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[0.9],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\",\"doc\":null,\"tags\":[]}},{\"id\":\"6ea307e2-93aa-4cdd-cc53-0fe234bdd645\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"green\"},\"line_color\":{\"value\":\"green\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"6ea307e2-93aa-4cdd-cc53-0fe234bdd645\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"16ca4d4e-69df-4fa1-cf2a-9b4b1287e209\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"green\"},\"line_color\":{\"value\":\"green\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"16ca4d4e-69df-4fa1-cf2a-9b4b1287e209\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"bc15cc70-ce9e-41f7-c15e-cf43818825f6\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"16ca4d4e-69df-4fa1-cf2a-9b4b1287e209\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\"},\"glyph\":{\"type\":\"Line\",\"id\":\"6ea307e2-93aa-4cdd-cc53-0fe234bdd645\"},\"selection_glyph\":null,\"id\":\"bc15cc70-ce9e-41f7-c15e-cf43818825f6\",\"tags\":[]}},{\"id\":\"f58a716c-0919-4ed1-ce81-4532f5591c71\",\"type\":\"DataRange1d\",\"attributes\":{\"sources\":[{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\"}},{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\"}},{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\"}},{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\"}}],\"id\":\"f58a716c-0919-4ed1-ce81-4532f5591c71\",\"tags\":[],\"doc\":null}},{\"id\":\"7d0e3034-00f9-433f-c303-8ee9f15f43eb\",\"type\":\"DataRange1d\",\"attributes\":{\"sources\":[{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\"}},{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\"}},{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\"}},{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\"}}],\"id\":\"7d0e3034-00f9-433f-c303-8ee9f15f43eb\",\"tags\":[],\"doc\":null}},{\"id\":\"3f6751a2-733b-4934-c931-1d1d86374e94\",\"type\":\"ToolEvents\",\"attributes\":{\"tags\":[],\"id\":\"3f6751a2-733b-4934-c931-1d1d86374e94\",\"geometries\":[],\"doc\":null}},{\"id\":\"916b9db1-7745-489e-c89f-1f690cc20823\",\"type\":\"BasicTickFormatter\",\"attributes\":{\"id\":\"916b9db1-7745-489e-c89f-1f690cc20823\",\"tags\":[],\"doc\":null}},{\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\",\"type\":\"BasicTicker\",\"attributes\":{\"num_minor_ticks\":5,\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\",\"tags\":[],\"doc\":null}},{\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\",\"type\":\"LinearAxis\",\"attributes\":{\"formatter\":{\"type\":\"BasicTickFormatter\",\"id\":\"916b9db1-7745-489e-c89f-1f690cc20823\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\"},\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"axis_label\":null,\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\",\"doc\":null,\"tags\":[]}},{\"id\":\"a49f8f88-172c-43fc-c36e-1e4be944b303\",\"type\":\"Grid\",\"attributes\":{\"dimension\":0,\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\"},\"id\":\"a49f8f88-172c-43fc-c36e-1e4be944b303\",\"doc\":null,\"tags\":[]}},{\"id\":\"4c92973d-d7fe-41e5-c1dd-b80d028b6905\",\"type\":\"BasicTickFormatter\",\"attributes\":{\"id\":\"4c92973d-d7fe-41e5-c1dd-b80d028b6905\",\"tags\":[],\"doc\":null}},{\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\",\"type\":\"BasicTicker\",\"attributes\":{\"num_minor_ticks\":5,\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\",\"tags\":[],\"doc\":null}},{\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\",\"type\":\"LinearAxis\",\"attributes\":{\"formatter\":{\"type\":\"BasicTickFormatter\",\"id\":\"4c92973d-d7fe-41e5-c1dd-b80d028b6905\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\"},\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"axis_label\":null,\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\",\"doc\":null,\"tags\":[]}},{\"id\":\"fc73a7e6-7404-4f02-cf0e-21bbb4c021ef\",\"type\":\"Grid\",\"attributes\":{\"dimension\":1,\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\"},\"id\":\"fc73a7e6-7404-4f02-cf0e-21bbb4c021ef\",\"doc\":null,\"tags\":[]}},{\"id\":\"3341b4ae-cbb7-4b06-cb09-af6b68284b67\",\"type\":\"PanTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"dimensions\":[\"width\",\"height\"],\"id\":\"3341b4ae-cbb7-4b06-cb09-af6b68284b67\",\"doc\":null,\"tags\":[]}},{\"id\":\"4790005a-184c-44cd-c430-d44d8e4487c9\",\"type\":\"WheelZoomTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"dimensions\":[\"width\",\"height\"],\"id\":\"4790005a-184c-44cd-c430-d44d8e4487c9\",\"doc\":null,\"tags\":[]}},{\"id\":\"851d1b8b-11ef-4d4f-cd56-15c4c5291a50\",\"type\":\"BoxZoomTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"851d1b8b-11ef-4d4f-cd56-15c4c5291a50\",\"tags\":[],\"doc\":null}},{\"id\":\"edb83bf6-05e3-44b6-c4a2-4de8e78dc2eb\",\"type\":\"PreviewSaveTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"edb83bf6-05e3-44b6-c4a2-4de8e78dc2eb\",\"tags\":[],\"doc\":null}},{\"id\":\"9ad64c8d-0dc1-490d-c9da-0dd82bac1e91\",\"type\":\"ResizeTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"9ad64c8d-0dc1-490d-c9da-0dd82bac1e91\",\"tags\":[],\"doc\":null}},{\"id\":\"88fe02b4-c305-4443-c4bc-51a48e8ab53c\",\"type\":\"ResetTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"88fe02b4-c305-4443-c4bc-51a48e8ab53c\",\"tags\":[],\"doc\":null}},{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"attributes\":{\"x_range\":{\"type\":\"DataRange1d\",\"id\":\"f58a716c-0919-4ed1-ce81-4532f5591c71\"},\"tool_events\":{\"type\":\"ToolEvents\",\"id\":\"3f6751a2-733b-4934-c931-1d1d86374e94\"},\"below\":[{\"type\":\"LinearAxis\",\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\"}],\"renderers\":[{\"type\":\"GlyphRenderer\",\"id\":\"b8bff12f-b404-43d6-c3ca-4404aedee39e\"},{\"type\":\"GlyphRenderer\",\"id\":\"dfc5d210-21c2-4941-c9e8-e315d3e43f11\"},{\"type\":\"GlyphRenderer\",\"id\":\"2ce6d563-645b-437e-c3e5-955c0880629e\"},{\"type\":\"GlyphRenderer\",\"id\":\"bc15cc70-ce9e-41f7-c15e-cf43818825f6\"},{\"type\":\"LinearAxis\",\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\"},{\"type\":\"Grid\",\"id\":\"a49f8f88-172c-43fc-c36e-1e4be944b303\"},{\"type\":\"LinearAxis\",\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\"},{\"type\":\"Grid\",\"id\":\"fc73a7e6-7404-4f02-cf0e-21bbb4c021ef\"}],\"above\":[],\"tools\":[{\"type\":\"PanTool\",\"id\":\"3341b4ae-cbb7-4b06-cb09-af6b68284b67\"},{\"type\":\"WheelZoomTool\",\"id\":\"4790005a-184c-44cd-c430-d44d8e4487c9\"},{\"type\":\"BoxZoomTool\",\"id\":\"851d1b8b-11ef-4d4f-cd56-15c4c5291a50\"},{\"type\":\"PreviewSaveTool\",\"id\":\"edb83bf6-05e3-44b6-c4a2-4de8e78dc2eb\"},{\"type\":\"ResizeTool\",\"id\":\"9ad64c8d-0dc1-490d-c9da-0dd82bac1e91\"},{\"type\":\"ResetTool\",\"id\":\"88fe02b4-c305-4443-c4bc-51a48e8ab53c\"}],\"doc\":null,\"right\":[],\"title\":\" 0 - 0% - 0 minutes\",\"extra_x_ranges\":{},\"left\":[{\"type\":\"LinearAxis\",\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\"}],\"y_range\":{\"type\":\"DataRange1d\",\"id\":\"7d0e3034-00f9-433f-c303-8ee9f15f43eb\"},\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"extra_y_ranges\":{},\"tags\":[]}}];\n",
        "  Bokeh.load_models(all_models);\n",
        "  var model = Bokeh.Collections(modeltype).get(modelid);\n",
        "  $(\"#3de3aa81-2bc2-4514-c55b-552a597be5ee\").html(''); // clear any previous plot in window_id\n",
        "  var view = new model.default_view({model: model, el: \"#3de3aa81-2bc2-4514-c55b-552a597be5ee\"});\n",
        "    } else {\n",
        "  load_lib(bokehjs_url, function() {\n",
        "      console.log(\"Bokeh: BokehJS plotting callback run at\", new Date())\n",
        "      var modelid = \"fb8d1727-b786-45ce-c531-9c54b5609220\";\n",
        "      var modeltype = \"Plot\";\n",
        "      var all_models = [{\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[2.3],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\",\"doc\":null,\"tags\":[]}},{\"id\":\"294bc1d6-012f-4c1b-ccc7-8ff65da8cfb9\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"red\"},\"line_color\":{\"value\":\"red\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"294bc1d6-012f-4c1b-ccc7-8ff65da8cfb9\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"8065d0d8-e099-4afe-ca1c-cd2fa32b7fc2\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"red\"},\"line_color\":{\"value\":\"red\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"8065d0d8-e099-4afe-ca1c-cd2fa32b7fc2\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"b8bff12f-b404-43d6-c3ca-4404aedee39e\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"8065d0d8-e099-4afe-ca1c-cd2fa32b7fc2\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\"},\"glyph\":{\"type\":\"Line\",\"id\":\"294bc1d6-012f-4c1b-ccc7-8ff65da8cfb9\"},\"selection_glyph\":null,\"id\":\"b8bff12f-b404-43d6-c3ca-4404aedee39e\",\"tags\":[]}},{\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[0.9],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\",\"doc\":null,\"tags\":[]}},{\"id\":\"50ab886e-2ffc-4562-c531-022fe52b2fa8\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"purple\"},\"line_color\":{\"value\":\"purple\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"50ab886e-2ffc-4562-c531-022fe52b2fa8\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"22748664-18c5-491a-c983-aa6519a2cf3d\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"purple\"},\"line_color\":{\"value\":\"purple\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"22748664-18c5-491a-c983-aa6519a2cf3d\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"dfc5d210-21c2-4941-c9e8-e315d3e43f11\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"22748664-18c5-491a-c983-aa6519a2cf3d\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\"},\"glyph\":{\"type\":\"Line\",\"id\":\"50ab886e-2ffc-4562-c531-022fe52b2fa8\"},\"selection_glyph\":null,\"id\":\"dfc5d210-21c2-4941-c9e8-e315d3e43f11\",\"tags\":[]}},{\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[2.3],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\",\"doc\":null,\"tags\":[]}},{\"id\":\"5009ef23-82c9-4a61-ca83-74682f17fda2\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"blue\"},\"line_color\":{\"value\":\"blue\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"5009ef23-82c9-4a61-ca83-74682f17fda2\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"cb31dc76-94d6-465b-c6f7-be2f9280a00e\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"blue\"},\"line_color\":{\"value\":\"blue\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"cb31dc76-94d6-465b-c6f7-be2f9280a00e\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"2ce6d563-645b-437e-c3e5-955c0880629e\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"cb31dc76-94d6-465b-c6f7-be2f9280a00e\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\"},\"glyph\":{\"type\":\"Line\",\"id\":\"5009ef23-82c9-4a61-ca83-74682f17fda2\"},\"selection_glyph\":null,\"id\":\"2ce6d563-645b-437e-c3e5-955c0880629e\",\"tags\":[]}},{\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\",\"type\":\"ColumnDataSource\",\"attributes\":{\"data\":{\"y\":[0.9],\"x\":[0]},\"column_names\":[\"y\",\"x\"],\"cont_ranges\":{},\"discrete_ranges\":{},\"selected\":[],\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\",\"doc\":null,\"tags\":[]}},{\"id\":\"6ea307e2-93aa-4cdd-cc53-0fe234bdd645\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"green\"},\"line_color\":{\"value\":\"green\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"6ea307e2-93aa-4cdd-cc53-0fe234bdd645\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"16ca4d4e-69df-4fa1-cf2a-9b4b1287e209\",\"type\":\"Line\",\"attributes\":{\"fill_alpha\":{\"units\":\"data\",\"value\":0.2},\"line_alpha\":{\"units\":\"data\",\"value\":1},\"doc\":null,\"size\":{\"units\":\"screen\",\"value\":10},\"fill_color\":{\"value\":\"green\"},\"line_color\":{\"value\":\"green\"},\"x\":{\"units\":\"data\",\"field\":\"x\"},\"id\":\"16ca4d4e-69df-4fa1-cf2a-9b4b1287e209\",\"y\":{\"units\":\"data\",\"field\":\"y\"},\"tags\":[]}},{\"id\":\"bc15cc70-ce9e-41f7-c15e-cf43818825f6\",\"type\":\"GlyphRenderer\",\"attributes\":{\"name\":null,\"nonselection_glyph\":{\"type\":\"Line\",\"id\":\"16ca4d4e-69df-4fa1-cf2a-9b4b1287e209\"},\"doc\":null,\"server_data_source\":null,\"data_source\":{\"type\":\"ColumnDataSource\",\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\"},\"glyph\":{\"type\":\"Line\",\"id\":\"6ea307e2-93aa-4cdd-cc53-0fe234bdd645\"},\"selection_glyph\":null,\"id\":\"bc15cc70-ce9e-41f7-c15e-cf43818825f6\",\"tags\":[]}},{\"id\":\"f58a716c-0919-4ed1-ce81-4532f5591c71\",\"type\":\"DataRange1d\",\"attributes\":{\"sources\":[{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\"}},{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\"}},{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\"}},{\"columns\":[\"x\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\"}}],\"id\":\"f58a716c-0919-4ed1-ce81-4532f5591c71\",\"tags\":[],\"doc\":null}},{\"id\":\"7d0e3034-00f9-433f-c303-8ee9f15f43eb\",\"type\":\"DataRange1d\",\"attributes\":{\"sources\":[{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"598729a1-0cf7-497a-c946-bb2b6c2bb0b4\"}},{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"90ac4a97-417d-4903-c98b-ed9a675d4918\"}},{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"dbcc3eaa-b727-47b6-c720-42663355dce0\"}},{\"columns\":[\"y\"],\"source\":{\"type\":\"ColumnDataSource\",\"id\":\"ae9bdc2a-8636-481f-c89f-c93e65dd2be2\"}}],\"id\":\"7d0e3034-00f9-433f-c303-8ee9f15f43eb\",\"tags\":[],\"doc\":null}},{\"id\":\"3f6751a2-733b-4934-c931-1d1d86374e94\",\"type\":\"ToolEvents\",\"attributes\":{\"tags\":[],\"id\":\"3f6751a2-733b-4934-c931-1d1d86374e94\",\"geometries\":[],\"doc\":null}},{\"id\":\"916b9db1-7745-489e-c89f-1f690cc20823\",\"type\":\"BasicTickFormatter\",\"attributes\":{\"id\":\"916b9db1-7745-489e-c89f-1f690cc20823\",\"tags\":[],\"doc\":null}},{\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\",\"type\":\"BasicTicker\",\"attributes\":{\"num_minor_ticks\":5,\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\",\"tags\":[],\"doc\":null}},{\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\",\"type\":\"LinearAxis\",\"attributes\":{\"formatter\":{\"type\":\"BasicTickFormatter\",\"id\":\"916b9db1-7745-489e-c89f-1f690cc20823\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\"},\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"axis_label\":null,\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\",\"doc\":null,\"tags\":[]}},{\"id\":\"a49f8f88-172c-43fc-c36e-1e4be944b303\",\"type\":\"Grid\",\"attributes\":{\"dimension\":0,\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"1abaa17f-bd72-4876-c83c-1f031df8990d\"},\"id\":\"a49f8f88-172c-43fc-c36e-1e4be944b303\",\"doc\":null,\"tags\":[]}},{\"id\":\"4c92973d-d7fe-41e5-c1dd-b80d028b6905\",\"type\":\"BasicTickFormatter\",\"attributes\":{\"id\":\"4c92973d-d7fe-41e5-c1dd-b80d028b6905\",\"tags\":[],\"doc\":null}},{\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\",\"type\":\"BasicTicker\",\"attributes\":{\"num_minor_ticks\":5,\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\",\"tags\":[],\"doc\":null}},{\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\",\"type\":\"LinearAxis\",\"attributes\":{\"formatter\":{\"type\":\"BasicTickFormatter\",\"id\":\"4c92973d-d7fe-41e5-c1dd-b80d028b6905\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\"},\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"axis_label\":null,\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\",\"doc\":null,\"tags\":[]}},{\"id\":\"fc73a7e6-7404-4f02-cf0e-21bbb4c021ef\",\"type\":\"Grid\",\"attributes\":{\"dimension\":1,\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"ticker\":{\"type\":\"BasicTicker\",\"id\":\"c1b533d8-3c18-4cdd-cc47-403647fe1018\"},\"id\":\"fc73a7e6-7404-4f02-cf0e-21bbb4c021ef\",\"doc\":null,\"tags\":[]}},{\"id\":\"3341b4ae-cbb7-4b06-cb09-af6b68284b67\",\"type\":\"PanTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"dimensions\":[\"width\",\"height\"],\"id\":\"3341b4ae-cbb7-4b06-cb09-af6b68284b67\",\"doc\":null,\"tags\":[]}},{\"id\":\"4790005a-184c-44cd-c430-d44d8e4487c9\",\"type\":\"WheelZoomTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"dimensions\":[\"width\",\"height\"],\"id\":\"4790005a-184c-44cd-c430-d44d8e4487c9\",\"doc\":null,\"tags\":[]}},{\"id\":\"851d1b8b-11ef-4d4f-cd56-15c4c5291a50\",\"type\":\"BoxZoomTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"851d1b8b-11ef-4d4f-cd56-15c4c5291a50\",\"tags\":[],\"doc\":null}},{\"id\":\"edb83bf6-05e3-44b6-c4a2-4de8e78dc2eb\",\"type\":\"PreviewSaveTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"edb83bf6-05e3-44b6-c4a2-4de8e78dc2eb\",\"tags\":[],\"doc\":null}},{\"id\":\"9ad64c8d-0dc1-490d-c9da-0dd82bac1e91\",\"type\":\"ResizeTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"9ad64c8d-0dc1-490d-c9da-0dd82bac1e91\",\"tags\":[],\"doc\":null}},{\"id\":\"88fe02b4-c305-4443-c4bc-51a48e8ab53c\",\"type\":\"ResetTool\",\"attributes\":{\"plot\":{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"subtype\":\"Figure\"},\"id\":\"88fe02b4-c305-4443-c4bc-51a48e8ab53c\",\"tags\":[],\"doc\":null}},{\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"type\":\"Plot\",\"attributes\":{\"x_range\":{\"type\":\"DataRange1d\",\"id\":\"f58a716c-0919-4ed1-ce81-4532f5591c71\"},\"tool_events\":{\"type\":\"ToolEvents\",\"id\":\"3f6751a2-733b-4934-c931-1d1d86374e94\"},\"below\":[{\"type\":\"LinearAxis\",\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\"}],\"renderers\":[{\"type\":\"GlyphRenderer\",\"id\":\"b8bff12f-b404-43d6-c3ca-4404aedee39e\"},{\"type\":\"GlyphRenderer\",\"id\":\"dfc5d210-21c2-4941-c9e8-e315d3e43f11\"},{\"type\":\"GlyphRenderer\",\"id\":\"2ce6d563-645b-437e-c3e5-955c0880629e\"},{\"type\":\"GlyphRenderer\",\"id\":\"bc15cc70-ce9e-41f7-c15e-cf43818825f6\"},{\"type\":\"LinearAxis\",\"id\":\"6b9474d1-8d60-4850-c806-e7a23f5fff7f\"},{\"type\":\"Grid\",\"id\":\"a49f8f88-172c-43fc-c36e-1e4be944b303\"},{\"type\":\"LinearAxis\",\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\"},{\"type\":\"Grid\",\"id\":\"fc73a7e6-7404-4f02-cf0e-21bbb4c021ef\"}],\"above\":[],\"tools\":[{\"type\":\"PanTool\",\"id\":\"3341b4ae-cbb7-4b06-cb09-af6b68284b67\"},{\"type\":\"WheelZoomTool\",\"id\":\"4790005a-184c-44cd-c430-d44d8e4487c9\"},{\"type\":\"BoxZoomTool\",\"id\":\"851d1b8b-11ef-4d4f-cd56-15c4c5291a50\"},{\"type\":\"PreviewSaveTool\",\"id\":\"edb83bf6-05e3-44b6-c4a2-4de8e78dc2eb\"},{\"type\":\"ResizeTool\",\"id\":\"9ad64c8d-0dc1-490d-c9da-0dd82bac1e91\"},{\"type\":\"ResetTool\",\"id\":\"88fe02b4-c305-4443-c4bc-51a48e8ab53c\"}],\"doc\":null,\"right\":[],\"title\":\" 0 - 0% - 0 minutes\",\"extra_x_ranges\":{},\"left\":[{\"type\":\"LinearAxis\",\"id\":\"f7004df0-ddf7-4d03-cdb8-4fa1de6e3bcf\"}],\"y_range\":{\"type\":\"DataRange1d\",\"id\":\"7d0e3034-00f9-433f-c303-8ee9f15f43eb\"},\"id\":\"fb8d1727-b786-45ce-c531-9c54b5609220\",\"extra_y_ranges\":{},\"tags\":[]}}];\n",
        "      Bokeh.load_models(all_models);\n",
        "      var model = Bokeh.Collections(modeltype).get(modelid);\n",
        "      $(\"#3de3aa81-2bc2-4514-c55b-552a597be5ee\").html(''); // clear any previous plot in window_id\n",
        "      var view = new model.default_view({model: model, el: \"#3de3aa81-2bc2-4514-c55b-552a597be5ee\"});\n",
        "  });\n",
        "    }\n",
        "});\n",
        "</script>\n",
        "<div class=\"plotdiv\" id=\"3de3aa81-2bc2-4514-c55b-552a597be5ee\"></div>\n"
       ],
       "metadata": {},
       "output_type": "display_data"
      },
      {
       "ename": "./cifar10_utils.lua:122: attempt to call method 'name' (a nil value)\nstack traceback:\n\t./cifar10_utils.lua:122: in function 'train_epoch'\n\t[string \"local timer = torch.Timer()...\"]:14: in main chunk\n\t[C]: in function 'xpcall'\n\t/home/cgel/torch/install/share/lua/5.1/itorch/main.lua:179: in function </home/cgel/torch/install/share/lua/5.1/itorch/main.lua:143>\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/poller.lua:75: in function 'poll'\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/impl/loop.lua:307: in function 'poll'\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/impl/loop.lua:325: in function 'sleep_ex'\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/impl/loop.lua:370: in function 'start'\n\t/home/cgel/torch/install/share/lua/5.1/itorch/main.lua:350: in main chunk\n\t[C]: in function 'require'\n\t[string \"arg={'/home/cgel/.ipython/profile_torch/secur...\"]:1: in main chunk",
       "evalue": "",
       "output_type": "pyerr",
       "traceback": [
        "./cifar10_utils.lua:122: attempt to call method 'name' (a nil value)\nstack traceback:\n\t./cifar10_utils.lua:122: in function 'train_epoch'\n\t[string \"local timer = torch.Timer()...\"]:14: in main chunk\n\t[C]: in function 'xpcall'\n\t/home/cgel/torch/install/share/lua/5.1/itorch/main.lua:179: in function </home/cgel/torch/install/share/lua/5.1/itorch/main.lua:143>\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/poller.lua:75: in function 'poll'\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/impl/loop.lua:307: in function 'poll'\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/impl/loop.lua:325: in function 'sleep_ex'\n\t/home/cgel/torch/install/share/lua/5.1/lzmq/impl/loop.lua:370: in function 'start'\n\t/home/cgel/torch/install/share/lua/5.1/itorch/main.lua:350: in main chunk\n\t[C]: in function 'require'\n\t[string \"arg={'/home/cgel/.ipython/profile_torch/secur...\"]:1: in main chunk"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "utils.net:forward(trainset.data[1]:cuda())\n",
      "itorch.image(trainset.data[1])\n",
      "itorch.image(utils.net:get(17).output[1])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {
        "png": {
         "height": 32,
         "width": 32
        }
       },
       "output_type": "display_data",
       "png": "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",
       "text": [
        "Console does not support images"
       ]
      },
      {
       "metadata": {
        "png": {
         "height": 28,
         "width": 84
        }
       },
       "output_type": "display_data",
       "png": "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",
       "text": [
        "Console does not support images"
       ]
      }
     ],
     "prompt_number": 68
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "utils.visualize_example = function (self, set, i)\n",
      "  print(classes[set.label[i]])\n",
      "  itorch.image(image.scale(set.data[i],90,90))\n",
      "  local predicted = utils.net:forward(set.data[i]:cuda())\n",
      "  predicted:exp()\n",
      "  local confidences, indices = torch.sort(predicted, true)\n",
      "  confidences = confidences[1]\n",
      "  for j = 1, predicted[1]:size(1) do\n",
      "    print(classes[indices[1][j]]..string.format(\"    %.3f%%\", confidences[j]*100))\n",
      "  end\n",
      "end"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 113
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "utils:visualize(2)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 114,
       "text": [
        "=======\t\n",
        "horse\t\n"
       ]
      },
      {
       "metadata": {
        "png": {
         "height": 90,
         "width": 90
        }
       },
       "output_type": "display_data",
       "png": "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",
       "text": [
        "Console does not support images"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 114,
       "text": [
        "horse    77,311%\t\n",
        "deer    22,395%\t\n",
        "dog    0,173%\t\n",
        "bird    0,074%\t\n",
        "frog    0,016%\t\n",
        "cat    0,009%\t\n",
        "truck    0,006%\t\n",
        "ship    0,006%\t\n",
        "airplane    0,006%\t\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 114,
       "text": [
        "automobile    0,006%\t\n",
        "=======\t\n",
        "horse\t\n"
       ]
      },
      {
       "metadata": {
        "png": {
         "height": 90,
         "width": 90
        }
       },
       "output_type": "display_data",
       "png": "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",
       "text": [
        "Console does not support images"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 114,
       "text": [
        "horse    100,000%\t\n",
        "deer    0,000%\t\n",
        "dog    0,000%\t\n",
        "bird    0,000%\t\n",
        "cat    0,000%\t\n",
        "airplane    0,000%\t\n",
        "truck    0,000%\t\n",
        "ship    0,000%\t\n",
        "frog    0,000%\t\n",
        "automobile    0,000%\t\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 114
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "itorch.image(utils.net:get(2).weight)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {
        "png": {
         "height": 80,
         "width": 30
        }
       },
       "output_type": "display_data",
       "png": "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",
       "text": [
        "Console does not support images"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 14,
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "itorch.image(utils.net:get(4).weight[1])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {
        "png": {
         "height": 80,
         "width": 30
        }
       },
       "output_type": "display_data",
       "png": "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",
       "text": [
        "Console does not support images"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 15,
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "torch.save(\"trained_models/\"..model_name .. \"bkup\", utils.net)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 13,
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}